ArticlePDF Available

The Spread of Behavior in an Online Social Network Experiment

Authors:

Abstract and Figures

How do social networks affect the spread of behavior? A popular hypothesis states that networks with many clustered ties and a high degree of separation will be less effective for behavioral diffusion than networks in which locally redundant ties are rewired to provide shortcuts across the social space. A competing hypothesis argues that when behaviors require social reinforcement, a network with more clustering may be more advantageous, even if the network as a whole has a larger diameter. I investigated the effects of network structure on diffusion by studying the spread of health behavior through artificially structured online communities. Individual adoption was much more likely when participants received social reinforcement from multiple neighbors in the social network. The behavior spread farther and faster across clustered-lattice networks than across corresponding random networks.
Content may be subject to copyright.
DOI: 10.1126/science.1185231
, 1194 (2010); 329Science et al.Damon Centola,
Experiment
The Spread of Behavior in an Online Social Network
This copy is for your personal, non-commercial use only.
. clicking herecolleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others
. herefollowing the guidelines can be obtained byPermission to republish or repurpose articles or portions of articles
(this information is current as of September 10, 2010 ):
The following resources related to this article are available online at www.sciencemag.org
http://www.sciencemag.org/cgi/content/full/329/5996/1194
version of this article at: including high-resolution figures, can be found in the onlineUpdated information and services,
http://www.sciencemag.org/cgi/content/full/329/5996/1194/DC1
can be found at: Supporting Online Material
found at: can berelated to this articleA list of selected additional articles on the Science Web sites
http://www.sciencemag.org/cgi/content/full/329/5996/1194#related-content
http://www.sciencemag.org/cgi/content/full/329/5996/1194#otherarticles
, 5 of which can be accessed for free: cites 20 articlesThis article
http://www.sciencemag.org/cgi/collection/sociology
Sociology : subject collectionsThis article appears in the following
registered trademark of AAAS. is aScience2010 by the American Association for the Advancement of Science; all rights reserved. The title CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on September 10, 2010 www.sciencemag.orgDownloaded from
G. Hunt, A. Miller, T. Olszewski, and P. Wagner for their
suggestions; M. Kosnik and A. Miller for reviews; and
M. Foote for verifying that my subsampling algorithms
were programmed correctly. Numerous contributors to the
Paleobiology Database made t his study possible, and
IamparticularlygratefultoM.Clapham,A.Hendy,and
W. Kiessling for recent contributions. Research described
here was funded by donations from anonymous private
individuals having no connection to it. This is Paleobiology
Database publication 117.
Supporting Online Material
www.sciencemag.org/cgi/content/full/329/5996/1191/DC1
Materials and Methods
Figs. S1 to S9
Tables S1 and S2
References
22 March 2010; accepted 30 June 2010
10.1126/science.1189910
The Spread of Behavior in an Online
Social Network Experiment
Damon Centola
How do social networks affect the spread of behavior? A popular hypothesis states that networks
with many clustered ties and a high degree of separation will be less effective for behavioral
diffusion than networks in which locally redundant ties are rewired to provide shortcuts across the
social space. A competing hypothesis argues that when behaviors require social reinforcement, a
network with more clustering may be more advantageous, even if the network as a whole has a
larger diameter. I investigated the effects of network structure on diffusion by studying the spread
of health behavior through artificially structured online communities. Individual adoption was
much more likely when participants received social reinforcement from multiple neighbors
in the social network. The behavior spread farther and faster across clustered-lattice networks than
across corresponding random networks.
Many behaviors spread through social
contact (13). As a result, the network
structure of who is connected to whom
can critically affect the extent to which a behav-
ior diffuses across a population (28). There are
two competing hypotheses about how network
structure affects diffusion. The strength of weak
tieshypothesis predicts that networks with
many long ties(e.g., small-worldtopologies)
will spread a social behavior farther and more
quickly than a network in which ties are highly
clustered (46). This hypothesis treats the spread
of behavior as a simple contagion, such as dis-
ease or information: A single contact with an
infectedindividual is usually sufficient to trans-
mit the behavior (2). The power of long ties is
that they reduce the redundancy of the diffusion
process by connecting people whose friends do
not know each other, thereby allowing a behavior
to rapidly spread to other areas of the network
(35). The ideal case for this lack of redundancy
is a randomnetwork, in which, in expectation
for a large population, each of an individuals
ties reaches out to different neighborhoods (4,9).
The other hypothesis states that, unlike disease,
social behavior is a complex contagion: People
usually require contact with multiple sources of
infectionbefore being convinced to adopt a be-
havior (2). This hypothesis predicts that because
clustered networks have more redundant ties,
which provide social reinforcement for adoption,
they will better promote the diffusion of behav-
iors across large populations (2,7). Despite the
scientific (6,7,10)andpractical(1,2,11)im-
portance of understanding the spread of behavior
through social networks, an empirical test of
these predictions has not been possible, because
it requires the ability to independently vary the
topological structure of a social network (12).
I tested the effects of network structure on
diffusion using a controlled experimental approach.
I studied the spread of a health behavior through
a network-embedded population by creating an
Internet-based health community, containing 1528
participants recruited from health-interest World
Wide Web sites (13).
Each participant created an anonymous online
profile, including an avatar, a user name, and a set
of health interests. They were then matched with
other participants in the studyreferred to as
health buddies”—as members of an online health
community. Participants could not contact their
health buddies directly, but they could receive
emails from the study informing them of their
health buddiesactivities. To preserve anonymity
and to prevent people from trying to identify
friends who may have also signed up for the study
(or from trying to contact health buddies outside
the context of the experiment), I blinded the
identifiers that people used. Participants made
decisions about whether or not to adopt a health
behavior based on the adoption patterns of their
health buddies. The health behavior used for this
study was the decision to register for an Internet-
based health forum, which offered access and rat-
ing tools for online health resources (13).
The health forum was not known (or acces-
sible) to anyone except participants in the ex-
periment. This ensured that the only sources of
encouragement that participants had to join the
forum were the signals that they received from their
health buddies. The forum was populated with ini-
tial ratings to provide content for the early adopters.
However, all subsequent content was contributed
by the participants who joined the forum.
Participants arriving to the study were randomly
assigned to one of two experimental conditions
a clustered-lattice network and a random network
that were distinguished only by the topological
structure of the social networks (Fig. 1). In the
clustered-latticenetwork condition, there was a
high level of clustering (5,6,13) created by re-
dundant ties that linked each nodes neighbors to
one another. The random network condition was
created by rewiring the clustered-lattice network
via a permutation algorithm based on the small-
worldnetwork model (6,1315). This ensured
that each node maintained the exact same number
of neighbors as in the clustered network (that is, a
homogeneous degree distribution), while simulta-
neously reducing clustering in the network and
eliminating redundant ties within and between
neighborhoods (4,6,14).
The network topologies were created before
the participants arrived, and the participants could
Sloan School of Management, Massachusetts Institute of
Technology, Cambridge, MA 02142, USA. E-mail: dcentola@
mit.edu
Fig. 1. Randomization of
participants to clustered-
lattice and random-
network conditions in a
single trial of this study
(N=128,Z=6).In
each condition, the black
node shows the focal
node of a neighborhood
to which an individual is
being assigned, and the
red nodes correspond to
that individualsneigh-
bors in the network. In
the clustered-lattice net-
work, the red nodes share
neighbors with each other, whereas in the random network they do not. White nodes indicate individuals who
are not connected to the focal node.
3 SEPTEMBER 2010 VOL 329 SCIENCE www.sciencemag.org1194
REPORTS
on September 10, 2010 www.sciencemag.orgDownloaded from
not alter the topology in which they were em-
bedded (e.g., by making new ties). In both condi-
tions, each participant was randomly assigned
to occupy a single node in one network. The
occupants of the immediately adjacent nodes in
the network (i.e., the network neighbors) consti-
tuted a participants health buddies (13). Each
node in a social network had an identical number
of neighbors as the other nodes in the network,
and participants could only see the immediate
neighbors to whom they were connected.
Consequently, the size of each participants
social neighborhood was identical for all par-
ticipants within a network and across conditions.
More generally, every aspect of a participants
experience before the initiation of the diffusion
dynamics was equivalent across conditions, and
the only difference between the conditions was
the pattern of connectedness of the social net-
works in which the participants were embedded.
Thus, any differences in the dynamics of diffu-
sion between the two conditions can be attri-
buted to the effects of network topology.
There are four advantages of this experi-
mental design over observational data. (i) The
present study isolates the effects of network
topology, independent of frequently co-occurring
factors such as homophily (3,16), geographic
proximity (17), and interpersonal affect (4,18),
which are easily conflated with the effects of
topological structure in observational studies
(2,3,11). (ii) I study the spread of a health-
related behavior that is unknown to the partici-
pants before the study (13), thereby eliminating
the effects of nonnetwork factors from the dif-
fusion dynamics, such as advertising, availability,
and pricing, which can confound the effects of
topology on diffusion when, for example, the
local structure of a social network correlates
with greater resources for learning about or
adopting an innovation (11 ,19). (iii) This study
eliminates the possibility for social ties to change
and thereby identifies the effects of network
structure on the dynamics of diffusion without
the confounding effects of homophilous tie
formation (1,20). (iv) Finally, this design allows
the same diffusion process to be observed
multiple times, under identical structural condi-
tions, thus allowing the often stochastic process of
individual adoption (21)tobestudiedinaway
that provides robust evidence for the effects of
network topology on the dynamics of diffusion.
I report the results from six independent trials
of this experimental design, each consisting of a
matched pair of network conditions. In each pair,
participants were randomized to either a clustered-
lattice network or a corresponding random net-
work (13). This yielded 12 independent diffusion
processes. Diffusion was initiated by selecting a
random seed node,which sent signals to its net-
work neighbors encouraging themto adopt a health-
related behaviornamely, registering for a health
forum Web site (13). Every time a participant
adopted the behavior (i.e., registered for the health
forum), messages were sent to her health buddies
inviting them to adopt. If a participant had mul-
tiple health buddies who adopted the behavior,
then she would receive multiple signals, one from
each neighbor. The more neighbors who adopted,
the more reinforcing signals a participant received.
The sequence of adoption decisions made by the
0 2 4 6 8 10 12 14
0
0.1
0.2
0.3
0.4
0.5
0.6
F
Time (Days)
Fraction Adopted
0 2 4 6 8
0
0.1
0.2
0.3
0.4
0.5
0.6
D
Time (Days)
Fraction Adopted
0 2 4 6 8 10 12 14
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.7 0.7
0.7 0.7
A
Time (Days)
Fraction Adopted
0 2 4 6 8 10 12 14
0
0.1
0.2
0.3
0.4
0.5
0.6
C
Time (Days)
Fraction Adopted
0 2 4 6 8 10 12 14 16 18 20 22
0
0.1
0.2
0.3
0.4
0.5
0.6
E
Time (Days)
Fraction Adopted
0 2 4 6 8 10 12 14
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
B
Time (Days)
Fraction Adopted
Fig. 2. Time series showing the adoption of a health behavior spreading through clustered-lattice (solid
black circles) and random (open triangles) social networks. Six independent trials of the study are
shown, including (A)N= 98, Z=6,(Bto D)N=128,Z=6,and(Eand F)N= 144, Z=8.Thesuccess
of diffusion was measured by the fraction of the total network that adopted the behavior. The speed of
the diffusion process was evaluated by comparing thetimerequiredforthebehaviortospreadtothe
greatest fraction reached by both conditions in each trial.
2 3 4
0.75
1.00
1.25
1.50
1.75
2.00
2.25
Reinforcing Signals
Hazard Ratio
Fig. 3. Hazard ratios for adoption for individuals
receiving two, three, and four social signals. The
hazard ratio gindicates that the likelihood of
adoption increases by a factor of gfor each ad-
ditional signal k, compared to the likelihood of
adoption from receiving k1 signals. The 95%
confidence intervals from the Cox proportional
hazards model are shown by error bars. The effect
of an additional signal on the likelihood of adop-
tion is significant if the 95% confidence interval
does not contain g=1(13).
www.sciencemag.org SCIENCE VOL 329 3 SEPTEMBER 2010 1195
REPORTS
on September 10, 2010 www.sciencemag.orgDownloaded from
members of each social network provides a pre-
cise time series of the spread of the behavior
through the population. It also provides an exact
record of the number of signals required for in-
dividuals to adopt the behavior. The starting time
(time = 0) for each diffusion process corresponds
to the instant when the seed node was activated
and the initial signals were sent. For each trial, the
diffusion process was allowed to run for 3 weeks
(~1.8 million seconds). To test for the possible
effects of population size (N) and degree (Z,the
number of health buddies each person had) on the
diffusion dynamics, I used three different versions
of the experiment: (i) N= 98, Z=6;(ii)N= 128,
Z=6;and(iii)N= 144, Z=8(13). The modest
range of population sizes tested and the corre-
spondingly narrow range of degrees were due to
the challenges of recruiting large numbers of peo-
ple simultaneously. Among the networks I used,
there were no effects of population size (13).
The results show that network structure has a
significant effect on the dynamics of behavioral
diffusion. Surprisingly, the topologies with greater
clustering and a larger diameter were much more
effective for spreading behavior. Figure 2 shows
the time series generated by the six indepen-
dent trials of the experiment. Adoption typically
spread to a greater fraction of the population in
the clustered networks (solid black circles) than
in the random networks (open triangles). On
average, the behavior reached 53.77% of the
clustered networks, whereas only 38.26% of
the population adopted in the random networks
(13). I also found that the behavior diffused
more quickly in the clustered networks than in
the random networks. The average rate of dif-
fusion in the clustered networks (0.2820 × 10
3
nodes/s) was more than four times faster than
that of the random condition (0.0643 × 10
3
nodes/s). Differences in both the success and
the rate of diffusion between network conditions
are statistically significant (P<0.01usingthe
Wilcoxon rank sumMann-Whitney Utest) (13).
The experimental findings were qualitatively
the same across different network and neighbor-
hood sizes. However, networks with a greater
degree (Z= 8) performed better than those with
a lower degree (Z= 6). Although this finding is
consistent with the hypothesis that more redun-
dant ties between neighborhoods can improve
the global spread of behavior, it may also indicate
that other topological features, such as degree
and density, are relevant factors affecting be-
havioral diffusion (2,7). This suggests impor-
tant avenues for future research.
At the individual level, the results (Fig. 3)
show that redundant signals significantly in-
creased the likelihood of adoption; social rein-
forcement from multiple health buddies made
participants much more willing to adopt the be-
havior. Figure 3 compares the baseline likelihood
of adoption after receiving one social signal to
the increased likelihood of adoption for nodes
receiving second, third, and fourth reinforcing
signals. Participants were significantly more likely
to adopt after receiving a second signal than
after receiving only one signal (P< 0.001 using
the Cox proportional hazards model). Receiving
a third signal also significantly increased the like-
lihood of adoption, but with a smaller marginal-
effect size (P< 0.05, Cox proportional hazards
model) (13). Additional signals had no significant
effect. This can be attributed to the attenuation of
the sample size as the number of signals increased.
A secondary, but important, issue related to
adoption is the level of commitment that individ-
uals have to a behavior once they have adopted it.
To investigate the effects of social reinforcement
on individualslevel of engagement with the
health forum, I compared the number of return
visits to the forum after registering, for adopters
grouped by the number of social signals that they
received (Fig. 4) (participants could not receive
additional signals once they had registered) .
Figure 4 shows pairwise comparisons of the
number of return visits for adopters receiving
only one signal (solid lines) versus those receiv-
ing two to five signals (dashed lines in panels A
to D, respectively). Though less than 15% of
adopters receiving one signal made a return visit
to the forum, more than 30% of participants re-
ceiving two signals made return visits, and 40%
of participants receiving three signals made at least
one return visit. Pairwise statistical comparisons
between group one and groups two through five
are all significant (P< 0.01 for all four compar-
isons, using the Kolmogorov-Smirnov test) (13),
indicating that participants who received more than
one social signal were significantly more likely to
return to the health forum than those who only
received a single signal. This suggests that there
was a significant effect of social reinforcement on
participantslevel of engagement with the adopted
behavior.
As with all experiments, design choices that
aided my control of the study also put constraints
on the behaviors that I could test. A key limitation
of my design is that, unlike in my experiment,
adopting a new health behavior is often extreme-
ly difficult in the real world. To adopt behaviors
such as getting a vaccination, going on a diet,
starting an exercise routine, or getting a screening,
people may be required to pay the costs of time,
deprivation, or even physical pain. Because of
this, I expect that the need for social reinforce-
ment would be greater for adopting these health
behaviors than it was for the behavior in my
study. Consequently, the diffusion of real-world
health behaviors may depend even more on
clustered-network structures than did the diffu-
sion dynamics reported in my results.
An additional constraint of my study was that
participants did not have any direct commu-
0 1 2 3 4 5
0.5
0.6
0.7
0.8
0.9
1
F(x)
C
Return Visits (x)
0 1 2 3 4 5
0.5
0.6
0.7
0.8
0.9
1
B
Return Visits (x)
F(x)
0 1 2 3 4 5
0.5
0.6
0.7
0.8
0.9
1
F(x)
A
Return Visits (x)
0 1 2 3 4 5
0.5
0.6
0.7
0.8
0.9
1
Return Visits (x)
D
F(x)
2 Signals
1 Signal
3 Signals
1 Signal
4 Signals
1 Signal
5 Signals
1 Signal
Fig. 4. Cumulative distribution functions of the number of return visits to the health forum (x)for
populations of adopters grouped by the number of signals that they received. Comparisons are shown
for adopters who received (A) one versus two signals, (B) one versus three signals, (C) one versus four
signals, and (D) one versus five signals. All pairwise comparisons between groups two through five with
each other showed no significant differences (P> 0.4 for all six comparisons, using the Kolmogorov-
Smirnov test) (13).
3 SEPTEMBER 2010 VOL 329 SCIENCE www.sciencemag.org
1196
REPORTS
on September 10, 2010 www.sciencemag.orgDownloaded from
nication with their health buddies or information
about their identities. This allowed me to isolate
the effects of network topology on the dynamics
of diffusion without the presence of confounding
variables. However, it also raises the question of
what the strength of the effects of network to-
pology would be when allowed to interact with
the effects of interpersonal relationships. An im-
portant assumption of this study is that the effects
of network topology will not be overwhelmed by
individualsexposure to other social factors. Pre-
vious studies have suggested that factors such as
homophily and strong interpersonal affect in
social ties can improve the diffusion of behav-
iors through social networks (3,18). In the real
world, these features of social relationships tend
to be highly correlated with the formation of
clustered social ties (3,22,23). Consequently, I
expect that these reinforcing factors would am-
plify the observed effects of clustered social
networks in promoting the diffusion of health
behaviors across a large population. However,
new experimental designs are required to test the
interaction effects of these variables (and other
variables such as gender, memory, and frequency
of interaction) on the spread of social behaviors.
Evidence in support of the strength of weak
tieshypothesis has suggested that networks with
high levels of local clustering and tightly knit
neighborhoods are inefficient for large-scale dif-
fusion processes (4,5,9). My findings show that,
not only is individual adoption improved by re-
inforcing signals that come from clustered social
ties (Fig. 3), but this individual-level effect also
translates into a system-level phenomenon where-
by large-scale diffusion can reach more people
and spread more quickly in clustered networks
than in random networks (Fig. 2). Whereas lo-
cally clustered ties may be redundant for simple
contagions, like information or disease (4,6,24),
they can be highly efficient for promoting behav-
ioral diffusion. On the basis of these findings, I
predict that public health interventions aimed at
the spread of new health behaviors (for instance,
improved diet, regular exercise, condom use, or
needle exchange) may do better to target clustered
residential networks rather than the casual contact
networks across which disease may spread very
quickly (25)particularly if the behaviors to be
diffused are highly complex (for instance, because
they are costly, difficult, or contravene existing
norms).
References and Notes
1. N. A. Christakis, J. H. Fowler, N. Engl. J. Med. 357, 370
(2007).
2. D. Centola, M. Macy, Am. J. Sociol. 113, 702 (2007).
3. E. M. Rogers, Diffusion of Innovations (Free Press,
New York, 1995).
4. M. Granovetter, Am. J. Sociol. 78, 1360 (1973).
5. D. J. Watts, Small Worlds: The Dynamics of Networks
Between Order and Randomness (Princeton Univ. Press,
Princeton, NJ, 1999).
6. D. J. Watts, S. H. Strogatz, Nature 393, 440
(1998).
7. D. Centola, V. Eguiluz, M. Macy, Phys. A 374, 449
(2007).
8. N. A. Christakis, J. H. Fowler, N. Engl. J. Med. 358, 2249
(2008).
9. M. E. J. Newman, J. Stat. Phys. 101, 819 (2000).
10. S. H. Strogatz, Nature 410, 268 (2001).
11. K. P. Smith, N. A. Christakis, Annu. Rev. Sociol. 34, 405
(2008).
12. M. Kearns, S. Suri, N. Montfort, Science 313, 824 (2006).
13. Materials and methods are available as supporting
material on Science Online.
14. S. Maslov, K. Sneppen, Science 296, 910 (2002).
15. S. Maslov, K. Sneppen, U. Alon, in Handbook of Graphs
and Networks, S. Bornholdt, H. G. Schuster, Eds.
(Wiley-VCH, Weinheim, Germany, 2003).
16. J. M. McPherson, L. Smith-Lovin, J. M. Cook, Annu. Rev.
Sociol. 27, 415 (2001).
17. P. Hedstrom, Am. J. Sociol. 99, 1157 (1994).
18. D. McAdam, R. Paulsen, Am. J. Sociol. 99, 640
(1993).
19. L. F. Berkman, I. Kawachi, Social Epidemiology (Oxford
Univ. Press, New York, 2000).
20. D. Centola, J. C. Gonzalez-Avella, V. Eguiluz, M. San Miguel,
J. Conf. Resl. 51, 905 (2007).
21. M. J. Salganik, P. S. Dodds, D. J. Watts, Science 311, 854
(2006).
22. G. Kossinets, D. J. Watts, Am. J. Sociol. 115, 405
(2009).
23. G. Kossinets, D. J. Watts, Science 311,88
(2006).
24. M. E. J. Newman, D. J. Watts, Phys. Rev. E 60, 7332
(1999).
25. P. D. Calvert et al., Nature 411, 90 (2001).
26. I thank A. van de Rijt and V. Eguiluz for helpful
comments, N. Christakis for useful discussion and
guidance, G. Pickard for programming assistance,
A. Wagner for developing the Healthy Lifestyle Network
Web site, T. Groves for all of the design work, and
J. Kreckler and K. Campbell at www.prevention.com and
G. Colditz of www.yourdiseaserisk.com for their assistance
recruiting participants. This work was supported in part
by the James S. McDonnell Foundation and the Robert
Wood Johnson Foundation.
Supporting Online Material
www.sciencemag.org/cgi/content/full/329/5996/1194/DC1
Materials and Methods
SOM Text
Figs. S1 to S7
Tables S1 to S3
References
26 November 2009; accepted 19 July 2010
10.1126/science.1185231
Human-Restricted Bacterial Pathogens
Block Shedding of Epithelial Cells by
Stimulating Integrin Activation
Petra Muenzner,
1
Verena Bachmann,
1
Wolfgang Zimmermann,
2
Jochen Hentschel,
1,3
Christof R. Hauck
1,4
*
Colonization of mucosal surfaces is the key initial step in most bacterial infections. One mechanism
protecting the mucosa is the rapid shedding of epithelial cells, also termed exfoliation, but
it is unclear how pathogens counteract this process. We found that carcinoembryonic antigen
(CEA)binding bacteria colonized the urogenital tract of CEA transgenic mice, but not of wild-type
mice, by suppressing exfoliation of mucosal cells. CEA binding triggered de novo expression of the
transforming growth factor receptor CD105, changing focal adhesion composition and activating
b
1
integrins. This manipulation of integrin inside-out signaling promotes efficient mucosal
colonization and represents a potential target to prevent or cure bacterial infections.
During colonization of mucosal surfaces,
incoming microbes must cope with mul-
tiple host defenses (1). One protective
mechanism of the mucosa in stratified and squa-
mous tissues is the accelerated turnover and
shedding of superficial epithelial cells, also re-
ferred to as exfoliation (24). Although in vitro
studies have suggested that microbes modulate
cell detachment (5,6), it is currently unknown how
successful mucosal pathogens deal with the ex-
foliation response in vivo. Neisseria gonorrhoeae,
a Gram-negative microorganism, causes one of
the most common sexually transmitted diseases
worldwide (7). Even though these bacteria can
induce the exfoliation of host cells upon contact
(811), they are able to establish themselves on
virtually every mucosal surface of the human
body.
To investigate how gonococci manage to col-
onize the urogenital mucosa efficiently, we per-
formed vaginal infection of female mice (12). In line
with the innate capacity of epithelial cells to re-
spond to this bacterial challenge, N. gonorrhoeae
triggered detachment of superficial epithelial cells
within 20 hours (Fig. 1A) and only small numbers
of gonococci could be re-isolated from wild-type
mice (Fig. 1B). Gonococci are adapted to humans
as their sole natural host. One of the host-specific
virulence traits that gonococci share with other
specialized mucosal colonizers, including Hae-
mophilus influenzae,Moraxella catarrhalis,and
N. meningitidis, is the ability to recognize hu-
1
Lehrstuhl Zellbiologie, Fachbereich Biologie, Universität
Konstanz, 78457 Konstanz, Germany.
2
Tumor Immunology
Laboratory, LIFE Center, Ludwig-Maximilians-Universität Mün-
chen, 81377 München, Germany.
3
EM Service, Fachbereich
Biologie, Universität Konstanz, 78457 Konstanz, Germany.
4
Konstanz Research School Chemical Biology, Universität
Konstanz, 78457 Konstanz, Germany.
*To whom correspondence should be addressed. E-mail:
christof.hauck@uni-konstanz.de
www.sciencemag.org SCIENCE VOL 329 3 SEPTEMBER 2010 1197
REPORTS
on September 10, 2010 www.sciencemag.orgDownloaded from
... In particular, SEP-G obtains a unified improvement in social network datasets, which differs from the performance in bioinformatics datasets. This performance divergence may be because SEP only relies on the network structure for hierarchical pooling, while the structural information in social network datasets is more redundant than that in bioinformatics datasets (Centola, 2010). It is worth noting that there are not any pooling methods suppress GIN in NCI1, or, put differently, pooling methods also do not show unified promotion in comparison with backbones. ...
Preprint
Full-text available
Following the success of convolution on non-Euclidean space, the corresponding pooling approaches have also been validated on various tasks regarding graphs. However, because of the fixed compression quota and stepwise pooling design, these hierarchical pooling methods still suffer from local structure damage and suboptimal problem. In this work, inspired by structural entropy, we propose a hierarchical pooling approach, SEP, to tackle the two issues. Specifically, without assigning the layer-specific compression quota, a global optimization algorithm is designed to generate the cluster assignment matrices for pooling at once. Then, we present an illustration of the local structure damage from previous methods in the reconstruction of ring and grid synthetic graphs. In addition to SEP, we further design two classification models, SEP-G and SEP-N for graph classification and node classification, respectively. The results show that SEP outperforms state-of-the-art graph pooling methods on graph classification benchmarks and obtains superior performance on node classifications.
... At present, there are four important perspectives of rumor analysis: source detection (16)(17)(18), propagation dynamics (19,20), fake information detection (21)(22)(23), containment, and intervention (24,25). A widely spread rumor or fake news can lead to reputation ruin (26), political consequences (27), and economic loss (28). Numerous scholars are dedicated to revealing the inherent driving mechanisms of rumors and putting forward effective strategies to prevent and control rumors (29)(30)(31)(32). ...
Article
Full-text available
Background In the early stage of the COVID-19 outbreak in China, several social rumors in the form of false news, conspiracy theories, and magical cures had ever been shared and spread among the general public at an alarming rate, causing public panic and increasing the complexity and difficulty of social management. Therefore, this study aims to reveal the characteristics and the driving factors of the social rumors during the COVID-19 pandemic.Methods Based on a sample of 1,537 rumors collected from Sina Weibo's debunking account, this paper first divided the sample into four categories and calculated the risk level of all kinds of rumors. Then, time evolution analysis and correlation analysis were adopted to study the time evolution characteristics and the spatial and temporal correlation characteristics of the rumors, and the four stages of development were also divided according to the number of rumors. Besides, to extract the key driving factors from 15 rumor-driving factors, the social network analysis method was used to investigate the driver-driver 1-mode network characteristics, the generation driver-rumor 2-mode network characteristics, and the spreading driver-rumor 2-mode characteristics.ResultsResearch findings showed that the number of rumors related to COVID-19 were gradually decreased as the outbreak was brought under control, which proved the importance of epidemic prevention and control to maintain social stability. Combining the number and risk perception levels of the four types of rumors, it could be concluded that the Creating Panic-type rumors were the most harmful to society. The results of rumor drivers indicated that panic psychology and the lag in releasing government information played an essential role in driving the generation and spread of rumors. The public's low scientific literacy and difficulty in discerning highly confusing rumors encouraged them to participate in spreading rumors.Conclusion The study revealed the mechanism of rumors. In addition, studies involving rumors on different emergencies and social platforms are warranted to enrich the findings.
... For example, Moussaid and colleagues (Moussaïd et al., 2017) found that evidence for judgment propagation was limited to a social distance of three to four degrees of separation. The extent of judgment propagation can also be limited by the variety of information sources (Centola, 2010), the social structure between senders and recipients (Aral & Walker, 2014), and participants' subjective perceptions of the information being judged (Moussaïd, Brighton, & Gaissmaier, 2015). ...
Article
Full-text available
Understanding the human factors governing effective information exchange is increasingly indispensable for the design of day to day human-computer systems. Moreover, effective information exchange becomes a matter of life or death during emergency egress. The complexity of an unknown environment and the unpredictable locations of hazards often prevent evacuees from identifying safe routes. Successful evacuations from locations impacted by fire or earthquakes may depend on user-generated information to increase the chance of collective survival. The present paper employed multi-user virtual reality experiments and an online survey to investigate the mechanisms underlying social influence and collective intelligence during emergencies. Our results demonstrate that information sharing helps to reduce evacuation time and trajectory length. Participants also shared more when given incentives or when there was a lack of knowledge in the public information pool. This work provides further indications of how collective intelligence can be promoted and deployed during emergencies.
... Traffic system is a typical self-driven, non-equilibrium and multiparticle system, which is a complex system with nonlinear interaction (Damon, 2010;Cao and Chugh, 2018). Ship behavior is influenced by the natural environment, infrastructure, traffic rules, and the marine traffic situation constituted by the surrounding ships, which have the characteristics of dynamics and uncertainty (Murray and Perera, 2021). ...
Article
Full-text available
With the continual development of modern transportation technology and artificial intelligence technology, how to recognize the complex phenomenon of ship behavior existing in maritime traffic has become a hot topic. Maritime traffic is a complex system, the emergence of ship behavior is a leading cause of traffic complexity, and make up the core ideas of this research. This research studies ship behavior from three aspects: ship individual behavior, ship-ship interaction and multi-ship behavior. According to the movement state attribute, the improved Social Force Model has been developed by considering of the interactive effects between ships. On that foundation, the complex network model has been built to analyze the emergence of multi-ship behavior in a macroscopic view. Through experimental analysis of ship behavior in different scenarios, the results show that the repulsive force between ships changes in the ship behavior dynamic model can express the dynamic characteristics of the ship. And structural entropy in marine traffic situation complex network has been proved to describe the maritime traffic system. As such, the framework proposed in this paper can provide a new perspective for further understanding and research of ship behavior.
Article
Full-text available
The rapid spread of conspiracy ideas associated with the recent COVID-19 pandemic represents a major threat to the ongoing and coming vaccination programs. Yet, the cognitive factors underlying the pandemic-related conspiracy beliefs are not well described. We hypothesized that such cognitive style is driven by delusion proneness, a trait phenotype associated with formation of delusion-like beliefs that exists on a continuum in the normal population. To probe this hypothesis, we developed a COVID-19 conspiracy questionnaire (CCQ) and assessed 577 subjects online. Their responses clustered into three factors that included Conspiracy, Distrust and Fear/Action as identified using principal component analysis. We then showed that CCQ (in particular the Conspiracy and Distrust factors) related both to general delusion proneness assessed with Peter’s Delusion Inventory (PDI) as well as resistance to belief update using a Bias Against Disconfirmatory Evidence (BADE) task. Further, linear regression and pathway analyses suggested a specific contribution of BADE to CCQ not directly explained by PDI. Importantly, the main results remained significant when using a truncated version of the PDI where questions on paranoia were removed (in order to avoid circular evidence), and when adjusting for ADHD- and autistic traits (that are known to be substantially related to delusion proneness). Altogether, our results strongly suggest that pandemic-related conspiracy ideation is associated with delusion proneness trait phenotype.
Article
The prevalence of obesity has increased substantially over the past 30 years. We performed a quantitative analysis of the nature and extent of the person-to-person spread of obesity as a possible factor contributing to the obesity epidemic. We evaluated a densely interconnected social network of 12,067 people assessed repeatedly from 1971 to 2003 as part of the Framingham Heart Study. The body-mass index was available for all subjects. We used longitudinal statistical models to examine whether weight gain in one person was associated with weight gain in his or her friends, siblings, spouse, and neighbors. Discernible clusters of obese persons (body-mass index [the weight in kilograms divided by the square of the height in meters], > or =30) were present in the network at all time points, and the clusters extended to three degrees of separation. These clusters did not appear to be solely attributable to the selective formation of social ties among obese persons. A person's chances of becoming obese increased by 57% (95% confidence interval [CI], 6 to 123) if he or she had a friend who became obese in a given interval. Among pairs of adult siblings, if one sibling became obese, the chance that the other would become obese increased by 40% (95% CI, 21 to 60). If one spouse became obese, the likelihood that the other spouse would become obese increased by 37% (95% CI, 7 to 73). These effects were not seen among neighbors in the immediate geographic location. Persons of the same sex had relatively greater influence on each other than those of the opposite sex. The spread of smoking cessation did not account for the spread of obesity in the network. Network phenomena appear to be relevant to the biologic and behavioral trait of obesity, and obesity appears to spread through social ties. These findings have implications for clinical and public health interventions.
Article
It is believed that almost any pair of people in the world can be connected to one another by a short chain of intermediate acquaintances, of typical length about six. This phenomenon, colloquially referred to as the “six degrees of separation,” has been the subject of considerable recent interest within the physics community. This paper provides a short review of the topic.
Article
The strength of weak ties is that they tend to be long - they connect socially distant locations, allowing information to diffuse rapidly. The authors test whether this "strength of weak ties" generalizes from simple to complex contagions. Complex contagions require social affirmation from multiple sources. Examples include the spread of high-risk social movements, avant garde fashions, and unproven technologies. Results show that as adoption thresholds increase, long ties can impede diffusion. Complex contagions depend primarily on the width of the bridges across a network, not just their length. Wide bridges are a characteristic feature of many spatial networks, which may account in part for the widely observed tendency for social movements to diffuse spatially.
Article
The authors investigate the origins of homophily in a large university community, using network data in which interactions, attributes, and affiliations are all recorded over time. The analysis indicates that highly similar pairs do show greater than average propensity to form new ties; however, it also finds that tie formation is heavily biased by triadic closure and focal closure, which effectively constrain the opportunities among which individuals may select. In the case of triadic closure, moreover, selection to "friend of a friend" status is determined by an analogous combination of individual preference and structural proximity. The authors conclude that the dynamic interplay of choice homophily and induced homophily, compounded over many "generations" of biased selection of similar individuals to structurally proximate positions, can amplify even a modest preference for similar others, via a cumulative advantagelike process, to produce striking patterns of observed homophily.
Article
Much empirical work in the social-movements literature has focused on the role of social ties in movement recruitment. Yet these studies have been plagued by a troubling theoretical and empirical imprecision. This imprecision stems from three sources. First, these studies are generally silent on the basic sociological dynamics that account for the reported findings. Second, movement scholars have generally failed to specify and test the precise dimensions of social ties that seem to account for their effects. Finally, most studies fail to acknowledge that individuals are embedded in many relationships that may expose the individual to conflicting pressures. This article seeks to address these shortcomings by means of an elaborated model of recruitment that is then used as a basis for examining the role of social ties in mediating individual recruitment to the 1964 Mississippi Freedom Summer Project.
Article
This article analyzes how distances and relations between actors are likely to influence the growth and spread of social movements. A formal theoretical model is developed that extends previous work on threshold models of collective behavior. Spatial distribution of a population influences the networks that are likely to emerge within the population; these networks, in turn, will influence the likely outcome of a mobilization effort. Key theoretical predictions are tested using data on the.founding of local union organizations in Sweden, 1890-1940. The empirical analyses show that contagious spatial processes were of considerable importance for the growth of the Swedish union movement, thus supporting the theoretical argument. The analyses presented in the article provide an alternative interpretation of density-dependent founding rates to the one offered by organizational ecologists.
Book
Getting an innovation adopted is difficult; a common problem is increasing the rate of its diffusion. Diffusion is the communication of an innovation through certain channels over time among members of a social system. It is a communication whose messages are concerned with new ideas; it is a process where participants create and share information to achieve a mutual understanding. Initial chapters of the book discuss the history of diffusion research, some major criticisms of diffusion research, and the meta-research procedures used in the book. This text is the third edition of this well-respected work. The first edition was published in 1962, and the fifth edition in 2003. The book's theoretical framework relies on the concepts of information and uncertainty. Uncertainty is the degree to which alternatives are perceived with respect to an event and the relative probabilities of these alternatives; uncertainty implies a lack of predictability and motivates an individual to seek information. A technological innovation embodies information, thus reducing uncertainty. Information affects uncertainty in a situation where a choice exists among alternatives; information about a technological innovation can be software information or innovation-evaluation information. An innovation is an idea, practice, or object that is perceived as new by an individual or an other unit of adoption; innovation presents an individual or organization with a new alternative(s) or new means of solving problems. Whether new alternatives are superior is not precisely known by problem solvers. Thus people seek new information. Information about new ideas is exchanged through a process of convergence involving interpersonal networks. Thus, diffusion of innovations is a social process that communicates perceived information about a new idea; it produces an alteration in the structure and function of a social system, producing social consequences. Diffusion has four elements: (1) an innovation that is perceived as new, (2) communication channels, (3) time, and (4) a social system (members jointly solving to accomplish a common goal). Diffusion systems can be centralized or decentralized. The innovation-development process has five steps passing from recognition of a need, through R&D, commercialization, diffusions and adoption, to consequences. Time enters the diffusion process in three ways: (1) innovation-decision process, (2) innovativeness, and (3) rate of the innovation's adoption. The innovation-decision process is an information-seeking and information-processing activity that motivates an individual to reduce uncertainty about the (dis)advantages of the innovation. There are five steps in the process: (1) knowledge for an adoption/rejection/implementation decision; (2) persuasion to form an attitude, (3) decision, (4) implementation, and (5) confirmation (reinforcement or rejection). Innovations can also be re-invented (changed or modified) by the user. The innovation-decision period is the time required to pass through the innovation-decision process. Rates of adoption of an innovation depend on (and can be predicted by) how its characteristics are perceived in terms of relative advantage, compatibility, complexity, trialability, and observability. The diffusion effect is the increasing, cumulative pressure from interpersonal networks to adopt (or reject) an innovation. Overadoption is an innovation's adoption when experts suggest its rejection. Diffusion networks convey innovation-evaluation information to decrease uncertainty about an idea's use. The heart of the diffusion process is the modeling and imitation by potential adopters of their network partners who have adopted already. Change agents influence innovation decisions in a direction deemed desirable. Opinion leadership is the degree individuals influence others' attitudes
Article
Similarity breeds connection. This principle - the homophily principle - structures network ties of every type, including marriage, friendship, work, advice, support, information transfer, exchange, comembership, and other types of relationship. The result is that people's personal networks are homogeneous with regard to many sociodemographic, behavioral, and intrapersonal characteristics. Homophily limits people's social worlds in a way that has powerful implications for the information they receive, the attitudes they form, and the interactions they experience. Homophily in race and ethnicity creates the strongest divides in our personal environments, with age, religion, education, occupation, and gender following in roughly that order. Geographic propinquity, families, organizations, and isomorphic positions in social systems all create contexts in which homophilous relations form. Ties between nonsimilar individuals also dissolve at a higher rate, which sets the stage for the formation of niches (localized positions) within social space. We argue for more research on: (a) the basic ecological processes that link organizations, associations, cultural communities, social movements, and many other social forms; (b) the impact of multiplex ties on the patterns of homophily; and (c) the dynamics of network change over time through which networks and other social entities co-evolve.
Article
Random links between otherwise distant nodes can greatly facilitate the propagation of disease or information, provided contagion can be transmitted by a single active node. However, we show that when the propagation requires simultaneous exposure to multiple sources of activation, called complex propagation, the effect of random links can be just the opposite; it can make the propagation more difficult to achieve. We numerically calculate critical points for a threshold model using several classes of complex networks, including an empirical social network. We also provide an estimation of the critical values in terms of vulnerable nodes.